Robust Multiple Responses Parameters Optimization Based On Fuzzy Theory

Intelligent manufacturing has put forward a new challenge to product and process optimization,monitoring and diagnosis.Multiple responses optimization design is an important technology of the quality management science,robust multiple responses optimization belongs to the scientific problems of process improvement and process design.The the dissertation focuses on researching robust multiple responses optimization in complex products manufacturing process by using fuzzy theory,taking account of the uncertainty in the design of response parameters,simultaneously carrying out the extending researches of fuzzy theory and robust multiple responses optimization.The following main contents are studied:(1)As an important generalization of fuzzy theory,the dissertation combines the interaction ideas with Bonferroni means and then extends them to intuitionistic fuzzy environments,proposes some intuitionstic fuzzy interaction Bonferroni means(IFIBMs).We obtain the mathematical formulas of the IFIBMs by mathematical induction,and discuss their properties.We investigate the properties of the IFIBMs,develop the multiple attributes decision making method based on the new Bonferroni mean under intuitionistic fuzzy environments by taking different generation function.The numerical examples show the effectiveness of the new methods.(2)The dissertation considers the unequal variance and model uncertainty of the responses,and takes account of the optimality and robustness of the solutions in multiple responses opatimization problems,proposes the multiple responses optimization approach based on the robust interactive desirability function.We incorporate the model uncertainty with the confidence intervals and get the robust solution with the optimization software.The new method is effective in generating a compromise solution which is more robust to uncertainties associated with model predictions,and thus improves the product quality,providing the theoretical basis for the product manufacturing.(3)Taking account of the uncertainty in the process of product design and production,the dissertation introduces the fuzzy theory to the robust multiple response optimization problems,we translate the feasible regions of multiple responses optimization problems into(?)-level sets and incorporate the model uncertainty with the confidence intervals simultaneously to ensure the robustness of the feasible regions.Then we develop the robust fuzzy programming approach to solve the multiple responses optimization (MRO) problems.Finally,we make comparisons of the algorithm in this dissertation with the conventional desirability function methods to illustrate the practically of the proposed method.(4)The dissertation constructs the triangular fuzzy desirability functions for different kinds of responses in multiple response optimization problems.The new desirability functions not only consider the traditional desirability value,robust desirability value,but also consider the fluctuation degree of the response values in the fixed confidence interval.The dissertation proposes the multiple responses optimization algorithm based on the robust triangular fuzzy desirability function and admissible orders.Finally,we make comparisons of the algorithm in this dissertation with the conventional desirability function method and robust desirability function in previous research to illustrate the advantages of the proposed method.We also consider the effects of different weights to the optimum solutions and make the sensitivity analysis of different methods to show the robustness of the final solutions.The dissertation extends the theories of fuzzy decision and the multi-response optimization design.It is helpful for engineering designers to deepen the understanding of product manufacturing process,providing the theoretical basis for enterprises to produce products that are much more robust.